AI stack · Agentic systems
Agents that plan, call tools, and stop.
Agentforce when the work lives in Salesforce. A code agent or SDK when it does not. MCP and A2A so the pieces talk without a custom API each time.
Founded 2022 · Founder in the Salesforce MVP Hall of Fame · Dubai HQ · Kochi studio
- 01Models · Claude, Gemini, OpenAI, Bedrock
- 02Agentic systems · Agentforce, CrewAI, ADK, A2A
- 03Agentic memory · Neo4j, Cognee, IndyKite
- 04Automation and voice · n8n, Retell
- 05Deployment and governance · identity, evals, policy, audit
On this layer
What we put in.
The honest no: if the work is a software factory with a named delivery date, we are the wrong partner.
Google ADK
Google Agent Development Kit on Vertex AI.
A2A Protocol
Agent2Agent interop across CrewAI, ADK, and Agentforce.
Claude Agent SDK
Multi-turn Claude agents that validate output before writing.
OpenAI Agents SDK
Typed agent steps: structured outputs and function calling.
LangGraph
Cyclic agent graphs with human approval gates.
Agentforce
Salesforce-native agents. Use when the work lives in the org.
CrewAI
Multi-agent automation that replaces brittle RPA scripts.
MuleSoft Agent Fabric
Orchestrate AI agents across your Salesforce ecosystem.
Model Context Protocol
Standard tool and data connectors for AI agents.
Where it runs
Agentforce, or a custom agent.
Agentforce fits when
A custom agent fits when
- The users work in Salesforce
- The seat is Slack, WhatsApp or another system
- The record lives in the org
- The record lives elsewhere
- Salesforce permissions already hold
- You need to own the model and the control plane
Where it shows up
The work this layer does.
Use cases where these tools carry the job. Each one names the stack.
AI IT Support Transformation
Ticket triage and knowledge agents on Vertex.
Quality assurance
Specialist agents for test design and defect clustering.
Data analytics
Agents that query governed datasets, not a raw warehouse dump.
Product development
Research and spec agents that stay inside Google Cloud.
AI Customer Service Transformation
Voice, chat, and CRM agents handing off without a custom API each time.
Supply chain
Planning agents that call finance or inventory specialists.
Compliance monitoring
A record of which agent called which agent, and why.
Legal document review
Contract and clause work the org-native agent should refuse.
Contract management
Lifecycle review with citations a reviewer can open.
Where the agent sits
Three desks, three runtimes.
Typical answers to “Agentforce or custom?”, from the seat and the record.
Example
Head of service, Dubai property developer
Today: The service team, the buyer records and the permissions all live in Salesforce.
After: Agentforce, inside the sharing model the team already has. No second place to govern.
Real estateExample
Fleet manager, Dubai last-mile courier
Today: Drivers report in WhatsApp, and the job record sits in the dispatch system.
After: A custom agent works from WhatsApp and reaches each system through governed MCP tools.
MCPExample
IT director, Abu Dhabi hospital group
Today: An Agentforce agent and a Python agent both run, and a coordinator copies output between them.
After: A2A lets them hand work to each other, so nobody copies between windows.
Healthcare
One request
How one request moves through the stack.
A customer asks a question on WhatsApp. Five stops later the record is right and someone can see why.
Surface
On WhatsApp, Slack, a portal or the phone. The agent acknowledges at once.
Gateway
The request runs as that person. The gateway decides which tools and models it may call.
Runtime
Agentforce if the work lives in Salesforce. A code agent if it does not.
Memory
The customer profile, the last conversation and the approved articles, with permissions attached.
Write
A queued write with retries. A timeout never posts twice. The trace is kept.
Questions
What buyers ask us.
- Agentforce or a custom agent?
- Agentforce when the records, the permissions and the people already live in Salesforce. A custom agent on an SDK when the seat is Slack, WhatsApp or another system. Many programs run both, with a control plane in front.
- Do we need MCP or A2A?
- MCP when an agent needs governed access to tools and data. A2A when agents from different vendors must hand work to each other. Neither is required for a first agent on one process.
- How do you stop an agent doing something wrong?
- Limits on steps and tools, an identity that inherits the user permissions, and a person on any write that commits the firm. Every run is logged so a reviewer can see why it acted.
The rest of the stack
One layer is not a system.
The plane this layer lives in, and the other four layers.
Runtime plane
Classify the step first. The framework is a consequence.
Models
The reasoning layer. We pick the model that fits the job, then deploy it so data stays where it should.
Agentic memory
What the agent remembers: graphs, identity, and long-term context so answers stay grounded.
Automation and voice
Workflows and voice agents that connect models to Salesforce and the rest of the stack.
Deployment and governance
The work that turns a pilot into a production system: identity, evaluations, policy and audit.
Before you pick a partner
Read these first. Then call anyone.
Our buyer guides. They apply whether or not you hire us.
Choosing an AI partner in the UAE
Five kinds of partner compared, the PDPL, DIFC, Arabic and WhatsApp questions, and red flags.
Choosing an AI implementation partner
Evaluation criteria, deployment models, governance fluency, and red flags.
AI partner evaluation framework
A five-part scorecard: depth, governance, integration, security, delivery.
Questions to ask an AI partner
Architecture, security, governance, pricing, and who is accountable.
AI partner vs in-house team
Hiring timelines, real costs, and dependency risk, scenario by scenario.
AI pilot to production
Which of the four ways your pilot stalled, and the sequence that moves it.
The brief
Send the situation.
Name the process. We tell you which layer is missing. The assessment is free: a short read of the org and a written note. If you go ahead, the first build week is a fixed AED/SAR 14,900 (about USD 4,050), credited in full when you sign the build within 60 days.